Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method comprising: providing an image of a wafer that includes a line of material, wherein the image has an x-axis and a y-axis perpendicular to the x-axis; correlating, using a processor, a potential defect against an x-axis pixel grey level intensity chart of the image; correlating, using the processor, the potential defect against a y-axis pixel grey level intensity chart of the image; determining, using the processor, a position of the potential defect relative to the line of material on the wafer along the x-axis and along the y-axis based on a point of the potential defect on the x-axis pixel grey level intensity chart and a point of the potential defect on the y-axis pixel grey level intensity chart, respectively, wherein the determining the position includes: interpolating a location of the line of material; interpolating a location of the potential defect; and determining a distance between the location of the line of material and the location of the potential defect; and classifying, using the processor, the potential defect as a defect of interest or a nuisance event based on the position of the potential defect, wherein the defect of interest is a non-zero distance apart from the line of material along both the x-axis and y-axis, and wherein the nuisance event is on the line of material along at least one of the x-axis or the y-axis.
In semiconductor manufacturing, detecting and classifying defects on wafers is critical for yield improvement. Defects can be genuine issues (defects of interest) or false alarms (nuisance events). This invention addresses the challenge of accurately distinguishing between these two types by analyzing wafer images to determine defect positions relative to material lines. The method involves capturing an image of a wafer containing a line of material, where the image has an x-axis and a perpendicular y-axis. A processor correlates potential defects against pixel grey level intensity charts along both axes. The processor then determines the defect's position relative to the material line by interpolating the line's location, interpolating the defect's location, and calculating the distance between them. The defect is classified as a defect of interest if it is separated from the line along both axes, indicating a genuine issue. If the defect lies on the line along at least one axis, it is classified as a nuisance event, reducing false positives. This approach improves defect detection accuracy in semiconductor inspection processes.
2. The method of claim 1 , wherein both the x-axis pixel grey level intensity chart and the y-axis pixel grey level intensity chart intersect the potential defect.
A method for detecting defects in an image involves analyzing pixel grey level intensity distributions along both the x-axis and y-axis of the image. The method generates a chart representing grey level intensity variations along the x-axis and another chart for the y-axis. These charts are used to identify potential defects by examining where the intensity distributions intersect with the defect. The intersection points on both charts indicate regions where the defect significantly alters the grey level intensity, making it detectable. This approach enhances defect detection accuracy by cross-referencing intensity changes in both horizontal and vertical directions. The method is particularly useful in quality control applications where precise defect localization is required, such as in semiconductor inspection, printed circuit board analysis, or material surface inspection. By comparing the intensity profiles along both axes, the method reduces false positives and improves the reliability of defect identification. The intersection of the x-axis and y-axis intensity charts with the defect ensures that only significant deviations are flagged, minimizing errors caused by noise or minor variations in the image. This technique is applicable to various imaging systems where grey level intensity analysis is used to detect anomalies.
3. The method of claim 1 , wherein both the x-axis pixel grey level intensity chart and the y-axis pixel grey level intensity chart include a threshold, wherein the threshold determines the line of material.
This invention relates to image processing techniques for determining the line of material in an object, particularly in applications where pixel grey level intensity is analyzed. The problem addressed involves accurately identifying the boundary or line of material within an image, which is crucial for quality control, defect detection, or material characterization in manufacturing or inspection processes. The method involves generating two separate pixel grey level intensity charts: one for the x-axis and one for the y-axis. Each chart represents the intensity distribution of pixels along its respective axis. A threshold is applied to both charts to determine the line of material. The threshold acts as a cutoff value that distinguishes between material and non-material regions in the image. By analyzing the intensity values relative to the threshold, the system can identify the boundary or line where the material is present. The method may also include preprocessing steps to enhance image quality, such as noise reduction or contrast adjustment, to improve the accuracy of the threshold-based analysis. The threshold can be dynamically adjusted based on the image characteristics or predefined criteria to ensure reliable detection of the material line. This approach is particularly useful in applications where precise material boundary identification is required, such as in semiconductor inspection, metal processing, or other industrial imaging tasks.
4. The method of claim 1 , further comprising identifying, using the processor, the potential defect in the image.
This invention relates to automated defect detection in images, particularly for identifying potential defects in visual data. The method involves processing an image to detect anomalies or irregularities that may indicate defects. The system uses a processor to analyze the image, applying computational techniques to identify regions or features that deviate from expected patterns. These deviations are flagged as potential defects, which can then be reviewed or further analyzed. The method may include preprocessing the image to enhance defect visibility, such as adjusting contrast or applying filters. The processor may also compare the image to reference data or templates to determine deviations. The defect identification step involves pattern recognition, statistical analysis, or machine learning models trained to detect anomalies. The system may output the identified defects with their locations and severity levels. This approach is useful in quality control, manufacturing, and inspection processes where automated defect detection improves efficiency and accuracy. The method ensures that potential defects are highlighted for further investigation, reducing human error and increasing reliability in defect detection.
5. The method of claim 1 , further comprising, determining, using the processor, a location of the line of material in the image.
This invention relates to automated inspection systems for detecting defects in materials, particularly for identifying lines of material in images and analyzing their properties. The system uses image processing techniques to capture and analyze images of a material surface, such as a metal sheet or fabric, to detect defects like cracks, tears, or irregularities. The method involves capturing an image of the material using an imaging device, such as a camera or scanner, and processing the image to identify a line of material within the image. The system then determines the location of the line by analyzing pixel data, edge detection, or pattern recognition algorithms. This location data can be used to assess the material's quality, detect defects, or guide further processing steps. The method may also include comparing the detected line to predefined standards or thresholds to determine if the material meets quality criteria. The system can be applied in manufacturing, quality control, or automated inspection processes where precise detection of material lines is critical. The invention improves defect detection accuracy and efficiency by automating the analysis of material lines in images.
6. The method of claim 1 , wherein the image is 32 pixels in the x-axis and 32 pixels in the y-axis.
This invention relates to image processing, specifically methods for handling small, fixed-size images. The technology addresses the need for efficient processing of compact image data, particularly in applications where memory and computational resources are limited. The method involves working with images that are precisely 32 pixels in width (x-axis) and 32 pixels in height (y-axis), ensuring uniformity in data handling. This fixed size simplifies storage, transmission, and processing tasks, as the system can optimize operations for this specific dimension. The approach is useful in embedded systems, real-time imaging, and low-power devices where predictable performance is critical. By standardizing the image size, the method reduces complexity in algorithms that analyze or manipulate the image data, such as compression, feature extraction, or pattern recognition. The invention may also include preprocessing steps to resize or crop input images to the 32x32 pixel format, ensuring compatibility with the processing pipeline. The fixed dimensions enable efficient memory allocation and faster execution times, making it suitable for applications like object detection, facial recognition, or machine vision in resource-constrained environments.
7. The method of claim 1 , wherein determining the position is based on a center of the potential defect.
A method for defect detection in materials or surfaces involves determining the position of a potential defect based on its center. The method uses imaging or sensing techniques to identify defects, such as cracks, voids, or irregularities, in materials like metals, composites, or coatings. The position of the defect is calculated by analyzing the center point of the detected anomaly, which helps in precisely locating the defect for further inspection or repair. This approach improves accuracy in defect mapping, ensuring that the exact location is recorded for quality control or maintenance purposes. The method may involve image processing, signal analysis, or other sensing technologies to detect and localize defects. By focusing on the center of the defect, the method reduces errors in positioning and enhances reliability in defect assessment. This technique is particularly useful in industries where precise defect localization is critical, such as aerospace, automotive, or manufacturing.
8. The method of claim 1 , further comprising generating, using the processor, the x-axis pixel grey level intensity chart and the y-axis pixel grey level intensity chart.
A system and method for analyzing pixel grey level intensity in digital images addresses the need for improved image processing techniques to enhance visual data interpretation. The invention involves capturing an image using an imaging device and processing the image data with a processor to extract pixel grey level intensity information. The processor generates a chart representing the grey level intensity distribution of pixels along the x-axis and another chart for the y-axis. These charts provide a visual representation of pixel intensity variations, enabling detailed analysis of image contrast, brightness, and other visual characteristics. The method further includes generating additional charts to compare intensity distributions across different axes, facilitating comprehensive image assessment. This approach enhances image quality evaluation, defect detection, and pattern recognition in various applications, including medical imaging, industrial inspection, and digital photography. The invention improves upon existing techniques by providing a more detailed and structured analysis of pixel intensity data, leading to more accurate and efficient image processing outcomes.
9. The method of claim 1 , wherein the line of material is a metal line.
A method for processing a metal line in a manufacturing or fabrication process involves controlling the deposition of material onto a substrate. The method includes positioning a substrate in a processing chamber, introducing a precursor gas into the chamber, and depositing a metal line onto the substrate using a chemical vapor deposition (CVD) or physical vapor deposition (PVD) technique. The deposition process is controlled by adjusting parameters such as gas flow rate, temperature, and pressure to ensure uniform and precise material deposition. The metal line may be part of an electronic or microelectronic device, such as a conductor in an integrated circuit. The method ensures high-quality deposition with minimal defects, improving the performance and reliability of the final product. The technique is particularly useful in semiconductor manufacturing, where precise control of metal line formation is critical for device functionality. The process may also include post-deposition treatments, such as annealing or etching, to refine the metal line's properties. This method addresses challenges in achieving consistent and defect-free metal line deposition, which is essential for advanced electronic applications.
10. A computer program product comprising a non-transitory computer readable storage medium having computer readable program embodied therewith, the computer readable program configured to carry out the method of claim 1 .
This invention relates to a computer program product for managing and analyzing data, particularly in systems where data integrity and efficient processing are critical. The problem addressed is the need for a reliable and automated way to process and validate data, ensuring accuracy and consistency across different datasets. The solution involves a non-transitory computer-readable storage medium containing program instructions that, when executed, perform a method for data processing. The method includes receiving input data, validating the data against predefined criteria, and generating output data based on the validation results. The validation process may involve checking for data completeness, consistency, and adherence to specified formats or rules. The program may also include error handling mechanisms to address discrepancies in the input data, such as flagging errors or correcting them automatically where possible. The output data may be formatted for further analysis or integration into other systems. The invention ensures that data is processed efficiently and accurately, reducing the risk of errors in downstream applications. The program is designed to be adaptable to various data types and validation requirements, making it suitable for a wide range of applications in industries such as finance, healthcare, and logistics.
11. A system comprising: a processor in electronic communication with an electronic storage medium and a wafer inspection tool, the processor configured to execute instructions that: correlate a potential defect against an x-axis pixel grey level intensity chart of the image of a wafer that includes a line of material, wherein the image has an x-axis and a y-axis perpendicular to the x-axis; correlate the potential defect against a y-axis pixel grey level intensity chart of the image; determine a position of the potential defect relative to the line of material on the wafer along the x-axis and along the y-axis based on a point of the potential defect on the x-axis pixel grey level intensity chart and a point of the potential defect on the y-axis pixel grey level intensity chart, respectively, wherein the determining the position includes: interpolating a location of the line of material; interpolating a location of the potential defect; and determining a distance between the location of the line of material and the location of the potential defect; and classify the potential defect as a defect of interest or a nuisance event based on the position of the potential defect, wherein the defect of interest is a non-zero distance apart from the line of material along both the x-axis and y-axis, and wherein the nuisance event is on the line of material along at least one of the x-axis or the y-axis.
The system is designed for defect classification in semiconductor wafer inspection, addressing the challenge of distinguishing true defects from nuisance events that do not affect wafer quality. The system uses a processor connected to an electronic storage medium and a wafer inspection tool to analyze images of wafers containing lines of material. The processor correlates potential defects against x-axis and y-axis pixel grey level intensity charts derived from the wafer image. By interpolating the location of the line of material and the potential defect, the system determines the defect's position relative to the line along both axes. The distance between the defect and the line is calculated to classify the defect as either a defect of interest or a nuisance event. A defect of interest is defined as one that is separated from the line by a non-zero distance in both the x and y directions, indicating a genuine defect. A nuisance event is identified when the defect lies on the line along at least one axis, suggesting it is not a true defect. This classification helps improve inspection accuracy by filtering out irrelevant events, reducing false positives, and enhancing semiconductor manufacturing yield.
12. The system of claim 11 , wherein both the x-axis pixel grey level intensity chart and the y-axis pixel grey level intensity chart include a threshold, wherein the threshold determines the line of material.
The invention relates to a system for analyzing material lines in an image, particularly in applications such as semiconductor inspection or material defect detection. The system addresses the challenge of accurately identifying material lines by analyzing pixel grey level intensity distributions along both the x-axis and y-axis of an image. Each axis includes a grey level intensity chart with a threshold value, which is used to determine the position of the material line. The threshold helps distinguish between the material line and background noise or other features in the image. The system likely processes the image to generate these intensity charts, applies the threshold to identify the line, and may use this information for further analysis, such as defect detection or quality control. The invention improves upon prior methods by providing a more precise and automated way to detect material lines based on pixel intensity variations. The system may be part of a larger inspection or measurement tool, ensuring accurate and consistent results in industrial or scientific applications.
13. The system of claim 11 , wherein the processor is further configured to execute instructions that identify the potential defect in the image.
The system is designed for automated defect detection in images, addressing the challenge of accurately identifying defects in visual data without manual intervention. The system processes images to detect potential defects by analyzing visual patterns, anomalies, or deviations from expected standards. It includes a processor configured to execute instructions that perform image analysis, such as pattern recognition, edge detection, or statistical analysis, to identify defects. The processor may also apply machine learning models trained on labeled defect data to improve detection accuracy. Additionally, the system may include a user interface for displaying detected defects and allowing user feedback to refine detection algorithms. The system may further integrate with manufacturing or quality control workflows to flag defective items for review or rejection. The defect detection process may involve comparing the image to a reference template or using historical data to determine whether a defect is present. The system aims to enhance efficiency and accuracy in defect detection across various industries, including manufacturing, healthcare, and automotive.
14. The system of claim 11 , wherein both the x-axis pixel grey level intensity chart and the y-axis pixel grey level intensity chart intersect the potential defect.
A system for defect detection in imaging systems analyzes pixel grey level intensity distributions to identify potential defects. The system generates an x-axis pixel grey level intensity chart and a y-axis pixel grey level intensity chart, each representing the distribution of grey level intensities along respective axes of an image. These charts are used to visualize and quantify variations in pixel intensity, which may indicate defects. The system further includes a defect detection module that processes these charts to determine whether a potential defect intersects both the x-axis and y-axis intensity distributions. By analyzing the intersection of these distributions, the system can more accurately identify and localize defects in the image. The intersection analysis helps distinguish between true defects and noise or artifacts, improving defect detection accuracy. This approach is particularly useful in high-resolution imaging applications where precise defect identification is critical, such as in semiconductor inspection, medical imaging, or industrial quality control. The system enhances defect detection by leveraging multi-axis intensity analysis to reduce false positives and improve detection reliability.
15. The system of claim 11 , wherein the processor is further configured to execute instructions that generate the x-axis pixel grey level intensity chart and the y-axis pixel grey level intensity chart.
A system for analyzing pixel grey level intensity in digital images processes image data to generate visual representations of pixel intensity distributions. The system includes a processor configured to execute instructions that generate an x-axis pixel grey level intensity chart and a y-axis pixel grey level intensity chart. These charts display the distribution of grey level intensities across pixels in an image, allowing for detailed analysis of image characteristics. The x-axis chart represents intensity values along the horizontal axis, while the y-axis chart represents intensity values along the vertical axis. This dual-axis approach provides a comprehensive view of pixel intensity variations, aiding in tasks such as image quality assessment, defect detection, and pattern recognition. The system may also include additional components for capturing or storing image data, ensuring compatibility with various imaging applications. By visualizing intensity distributions, the system helps identify inconsistencies, enhance image processing algorithms, and improve overall image analysis accuracy.
16. The system of claim 11 , wherein the line of material is a metal line.
The invention relates to a system for processing a line of material, particularly in manufacturing or fabrication environments where precise handling and manipulation of materials are required. The system addresses challenges in accurately controlling the position, orientation, or movement of a material line, which can be critical for ensuring product quality, efficiency, and consistency in automated production processes. The system includes a material line, a support structure, and a control mechanism. The support structure holds the material line in a fixed or adjustable position, while the control mechanism regulates the movement or alignment of the material line relative to the support structure. This control can involve adjusting tension, guiding the line through a specific path, or ensuring proper alignment with other components in the system. In one embodiment, the material line is a metal line, which may be used in applications such as wire forming, metal fabrication, or electronic component assembly. The system ensures that the metal line is properly positioned and manipulated to meet precise manufacturing requirements, reducing defects and improving production efficiency. The control mechanism may include sensors, actuators, or feedback systems to monitor and adjust the line's position dynamically. The support structure may be adjustable to accommodate different line lengths, thicknesses, or materials, enhancing the system's versatility. The invention aims to provide a robust and adaptable solution for handling material lines in automated or semi-automated production environments.
17. The system of claim 11 , wherein the processor is incorporated in the wafer inspection tool.
The system relates to semiconductor wafer inspection, addressing the need for efficient and accurate defect detection during manufacturing. The system includes a processor that analyzes data from a wafer inspection tool to identify defects on semiconductor wafers. The processor is integrated directly into the inspection tool, enabling real-time processing and immediate feedback. This integration reduces latency and improves throughput by eliminating the need for external data transfer or processing delays. The system may also include a memory for storing inspection data and a display for visualizing defect information. The processor executes algorithms to detect, classify, and quantify defects, providing detailed insights into wafer quality. By incorporating the processor within the inspection tool, the system enhances manufacturing efficiency and reduces the risk of defects escaping detection. The technology is particularly useful in high-volume semiconductor production, where rapid and precise defect analysis is critical for yield optimization.
18. The system of claim 17 , wherein the wafer inspection tool generates the image.
A wafer inspection system includes a wafer inspection tool that generates an image of a semiconductor wafer. The system also includes a processing unit that receives the image and identifies defects on the wafer. The processing unit applies a machine learning model to classify the defects based on their characteristics, such as size, shape, and location. The system further includes a database that stores defect data, including historical defect patterns and classification results. The processing unit compares the identified defects with the stored defect data to determine if the defects match known patterns. If a match is found, the system retrieves the corresponding classification from the database. If no match is found, the system uses the machine learning model to classify the defect. The system then generates a report summarizing the defect classifications and their locations on the wafer. The report may include recommendations for corrective actions based on the defect types identified. The wafer inspection tool may use optical or electron microscopy to capture high-resolution images of the wafer surface. The machine learning model is trained on a dataset of labeled defect images to improve accuracy in classification. The system is designed to improve yield in semiconductor manufacturing by identifying and categorizing defects efficiently, reducing manual inspection time and improving defect analysis accuracy.
19. The system of claim 18 , wherein the wafer inspection tool is a scanning electron microscope.
A wafer inspection system is designed to detect defects on semiconductor wafers during manufacturing. The system includes a wafer inspection tool that scans the wafer surface to identify defects, a defect classification module that categorizes detected defects, and a defect analysis module that evaluates the impact of classified defects on wafer yield. The system also includes a user interface for displaying inspection results and a data storage module for recording defect data. The wafer inspection tool is a scanning electron microscope (SEM), which provides high-resolution imaging to detect sub-micron defects on the wafer surface. The SEM scans the wafer using an electron beam, capturing images that are processed to identify defects. The defect classification module uses machine learning algorithms to categorize defects based on their size, shape, and location. The defect analysis module assesses the impact of each defect on wafer yield, considering factors such as defect density and criticality. The user interface allows operators to view inspection results, adjust parameters, and generate reports. The data storage module stores defect data for historical analysis and process optimization. This system improves defect detection accuracy and reduces false positives, enhancing semiconductor manufacturing yield.
Unknown
March 24, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.